Comment 1
Comment 2
Reference Information:
Title: Estimating User’s Engagement from Eye-gaze Behaviors in Human-Agent Conversations
Authors: Yukiko, I. Nakano, Ryo Ishii
Venue: IUI
Summary:
In this paper the authors discuss how eye contact is an important part of conversation. People judge others facial expressions in order to see if they are currently interested in the conversation they are having. The authors of this paper propose a means of doing the same thing with a computer. The system scans the persons eye gaze behaviors in order to determine if they are currently engaged in the conversation they are having with the system. If they are not, then the system will ask questions in order to make the user more engaged in the conversation.
Discussion:
In my opinion, this system has tremendous potential. The first possibility that comes to mind is adapting this system for students in order to tell when they are actually engaged in studying instead of doing something else or simply skimming. I do question if certain people could have strange facial expressions that the system has trouble reading.
I agree, the field of eye tracking has amazing potential. I hope that future work is done in this area.
ReplyDeleteI have to agree with you. This is a really interesting idea. I am kind of curious as to how they establish a baseline for how interested you are, though. Otherwise, if you start off disinterested, it might not work as well as if you start off interested.
ReplyDeleteI'd really like to try this out and see how accurate it is. Seems like a great system!
ReplyDelete